Papers by Ananth Ranganathan
Abstract We propose a new method for learning probabilistic part-based models of objects using on... more Abstract We propose a new method for learning probabilistic part-based models of objects using only a limited number of positive examples. The parts correspond to HOG bundles, which are groupings of HOG features. Each part model is supplemented by an appearance model, which captures the global appearance of the object by using bags of words of PHOW features.
Abstract Smoothing approaches to the Simultaneous Localization and Mapping (SLAM) problem in robo... more Abstract Smoothing approaches to the Simultaneous Localization and Mapping (SLAM) problem in robotics are superior to the more common filtering approaches in being exact, better equipped to deal with non-linearities, and computing the entire robot trajectory. However, while filtering algorithms that perform map updates in constant time exist, no analogous smoothing method is available.
Abstract—Self-organization is one of the few theories that can explain significant aspects of dev... more Abstract—Self-organization is one of the few theories that can explain significant aspects of developmental neuroscience. Within the brain itself, various spatially organized regions, or maps, exist that emerge dynamically. Theories and models that use self-organization have been successful at explaining such phenomena, and while these are not conclusive proof, they provide strong evidence in favor of self-organized mechanisms in the brain.
We present a new Gaussian Process inference algorithm, called Online Sparse Matrix Gaussian Proce... more We present a new Gaussian Process inference algorithm, called Online Sparse Matrix Gaussian Processes (OSMGP), and demonstrate its merits with a few vision applications. The OSMGP is based on the observation that for kernels with local support, the Gram matrix is typically sparse. Maintaining and updating the sparse Cholesky factor of the Gram matrix can be done efficiently using Givens rotations. This leads to an exact, online algorithm whose update time scales linearly with the size of the Gram matrix.
Smoothing approaches to the Simultaneous Localization and Mapping (SLAM) problem in robotics are ... more Smoothing approaches to the Simultaneous Localization and Mapping (SLAM) problem in robotics are superior to the more common filtering approaches in being exact, better equipped to deal with non-linearities, and computing the entire robot trajectory. However, while filtering algorithms that perform map updates in constant time exist, no analogous smoothing method is available.
This paper aims to present a structured variational inference algorithm for switching linear dyna... more This paper aims to present a structured variational inference algorithm for switching linear dynamical systems (SLDSs) which was initially introduced by Pavlovic and Rehg. Starting with the need for the variational approach, we proceed to the derivation of the generic (model-independent) variational update formulas which are obtained under the mean field assumption. This leads us to the derivation of an approximate variational inference algorithm for an SLDS.
A system and method are disclosed for learning a random multinomial logit (RML) classifier and ap... more A system and method are disclosed for learning a random multinomial logit (RML) classifier and applying the RML classifier for scene segmentation. The system includes an image textonization module, a feature selection module and a RML classifier. The image textonization module is configured to receive an image training set with the objects of the images being pre-labeled. The image textonization module is further configured to generate corresponding texton images from the image training set.
Abstract Categorizing areas such as rooms and corridors using a discrete set of labels has been o... more Abstract Categorizing areas such as rooms and corridors using a discrete set of labels has been of long-standing interest to the robotics community. A map with labels such as kitchen, lab, copy room etc provides a basic amount of semantic information that can enable a robot to perform a number of tasks specified in human-centric terms rather than just map coordinates. In this work, we propose a method to label areas in a pre-built map using information from camera images.
Abstract We illustrate the relationship between message passing algorithms in GMRFs and matrix fa... more Abstract We illustrate the relationship between message passing algorithms in GMRFs and matrix factorization algorithms. Specifically, we show that message passing on trees is equivalent to Gaussian elimination, while Loopy Belief Propagation is equivalent to Gauss-Seidel relaxation. Similarly, recently introduced message passing algorithms such as the Extended Message Passing, and the Embedded Subgraphs algorithm are also shown to be equivalent to commonly known matrix methods.
Insects are nature's engineering marvels. They can perform marvellous feats such as jumping many ... more Insects are nature's engineering marvels. They can perform marvellous feats such as jumping many times their own body length, lifting many times their own body weight, moving around even with multiple missing limbs and doing all this with amazing efficiency. It is little wonder then that roboticists have been attempting to replicate nature's design feat in robots. The primary thrust in the area of arthropod robotics has been locomotion.
Abstract This note aims to give a general overview of variational inference on graphical models. ... more Abstract This note aims to give a general overview of variational inference on graphical models. Starting with the need for the variational approach, we proceed to the derivation of the Variational Bayes EM algorithm that creates distributions on the hidden variables in a graphical model. This leads us to the Variational message Passing algorithm for conjugate exponential families, which is shown to result in a set of updates for the parameters of the distributions involved.
PDRRTs: Integrating graph-based and cell-based planning
Abstract Motion-planning problems can be solved by discretizing the continuous configuration spac... more Abstract Motion-planning problems can be solved by discretizing the continuous configuration space, for example with graph-based or cell-based techniques. We study rapidly exploring random trees (RRTs) as an example of graph-based techniques and the parti-game method as an example of cell-based techniques. We then propose parti-game directed RRTs (PDRRTs) as a novel technique that combines them.
We present an application of Bayesian modeling and inference to topological mapping in robotics. ... more We present an application of Bayesian modeling and inference to topological mapping in robotics. This is a potentially difficult problem due to (a) the combinatorial nature of the state space, and (b) perceptual aliasing by which two different landmarks in the environment can appear similar to the robot's sensors. Hence, this presents a challenging approximate inference problem, complicated by the fact that the form of the prior on topologies is far from obvious.
Having a map of the environment is almost an essential pre-requisite for robots to perform any ta... more Having a map of the environment is almost an essential pre-requisite for robots to perform any task requiring mobility. As robots enter everyday life via tasks as diverse as vacuum cleaning, nursing, and military transport, the requirement for maps that can enable mobility is obvious. Tasks such as surveying a disaster site during a search and rescue operation also require a map so that the location of possible victims or hazardous areas can be communicated.
Abstract Pose estimation of outdoor robots presents some distinct challenges due to the various u... more Abstract Pose estimation of outdoor robots presents some distinct challenges due to the various uncertainties in the robot sensing and action. In particular, global positioning sensors of outdoor robots do not always work perfectly, causing large drift in the location estimate of the robot. To overcome this common problem, we propose a new approach for global localization using place recognition.
Abstract We present a new Gaussian process (GP) inference algorithm, called online sparse matrix ... more Abstract We present a new Gaussian process (GP) inference algorithm, called online sparse matrix Gaussian processes (OSMGP), and demonstrate its merits by applying it to the problems of head pose estimation and visual tracking. The OSMGP is based upon the observation that for kernels with local support, the Gram matrix is typically sparse. Maintaining and updating the sparse Cholesky factor of the Gram matrix can be done efficiently using Givens rotations.
Abstract We describe a system for generating coherent movies from a collection of unedited videos... more Abstract We describe a system for generating coherent movies from a collection of unedited videos. The generation process is guided by one or more input keyframes, which determine the content of the generated video. The basic mechanism involves similarity analysis using the histogram intersection function. The function is applied to spatial pyramid histograms computed on the video frames in the collection using Dense SIFT features.
Abstract We present a method to reconstruct indoor environments from stereo image pairs, suitable... more Abstract We present a method to reconstruct indoor environments from stereo image pairs, suitable for the navigation of robots. To enable a robot to navigate solely using visual cues it receives from a stereo camera, the depth information needs to be extracted from the image pairs and combined into a common representation. The initially determined raw depthmaps are fused into a two level heightmap representation which contains a floor and a ceiling height level.
Abstract Automatic detection of landmarks, usually special places in the environment such as gate... more Abstract Automatic detection of landmarks, usually special places in the environment such as gateways, for topological mapping has proven to be a difficult task. We present the use of Bayesian surprise, introduced in computer vision, for landmark detection. Further, we provide a novel hierarchical, graphical model for the appearance of a place and use this model to perform surprise-based landmark detection.
Abstract We present a novel algorithm for topological mapping, which is the problem of finding th... more Abstract We present a novel algorithm for topological mapping, which is the problem of finding the graph structure of an environment from a sequence of measurements. Our algorithm, called Online Probabilistic Topological Mapping (OPTM), systematically addresses the problem by constructing the posterior on the space of all possible topologies given measurements. With each successive measurement, the posterior is updated incrementally using a Rao—Blackwellized particle filter.
Uploads
Papers by Ananth Ranganathan